Hi Sebastian,
Estimation with serially sampled DNA sequences is possible in DPPDiv, sort of. I implemented this, but it has not been tested. I suggest if you're interested in estimating divergence times with this kind of dataset, you might want to consider using BEAST2 or RevBayes since both have more machinery in place for that kind of analysis. However, BEAST2 does not have the DPP model on branch rates. If you are very interested in using that model for your dataset, it should be fairly straightforward in RevBayes, I just need to finish the testing and tutorial.
Cheers,
Tracy